the result?

[yyfyan]

Hi,Good job.
I used your trained model, VOC12_scenes_16000.pth, to test,But get the lower mIou 66.35%，not get your mIou result.

[isht7]

Would you mind sharing screenshot of your evaluation code used here?

[yyfyan]

I used the evalpyt2.py
evalpyt2.py.zip

[yyfyan]

and when read the train.py，I found the flip augment is not used,becauses you only flip the image ,not use the double images(origali image and flip image).

[marvis]

@yyfyan I got exactly the same result as yours.

evalpyt.py: pytorch 1 0.7467628417262675
evalpyt2.py: pytorch 1 Mean iou = 0.6635553130340005

Have you solved this problem?